Kappa coefficient

Summary

Doc_BoxAlgorithm_KappaCoefficient.png
  • Plugin name : Kappa coefficient
  • Version : 0.1
  • Author : Serrière Guillaume
  • Company : Inria
  • Short description : Compute the kappa coefficient for the classifier.
  • Documentation template generation date : Nov 2 2017

Description

The box computes kappa coefficient for a classifier.

This box computes the Cohen kappa coefficient that allows to compare the accordance of two classifiers (https://en.wikipedia.org/wiki/Cohen%27s_kappa), The box compares the results of the classifier (second input) to the 100% match classifier (first input).

The result is streamed on the first output, and is displayed in real time in a standalone visualization.

Inputs

1. Expected stimulations

The first input receives the expected stimulation.

  • Type identifier : Stimulations (0x6f752dd0, 0x082a321e)

2. Found stimulations

This input receives the stimualtions found by the classifier.

  • Type identifier : Stimulations (0x6f752dd0, 0x082a321e)

Outputs

1. Confusion Matrix

This output contains the current value of the Kappa coefficient.

  • Type identifier : Streamed matrix (0x544a003e, 0x6dcba5f6)

Settings

Each setting except the first one corresponds to the stimulation code of a class. A stimulation must be unique.

1. Number of classes

This setting indicates the amount of classes handled by the box. This setting will change the amount of setting.

  • Type identifier : Integer (0x007deef9, 0x2f3e95c6)
  • Default value : [ 2 ]

2. Stimulation of class 1

This setting indicates the stimulation corresponding to the first class.

  • Type identifier : Stimulation (0x2c132d6e, 0x44ab0d97)
  • Default value : [ OVTK_StimulationId_Label_01 ]

3. Stimulation of class 2

This setting indicates the stimulation corresponding to the second class.

  • Type identifier : Stimulation (0x2c132d6e, 0x44ab0d97)
  • Default value : [ OVTK_StimulationId_Label_02 ]

Examples

Miscellaneous